Role card
Semantic platform operator: the functional role
Who runs the graphs, the inference, the semantic view and the vector index — and keeps them in step with the approved model?
A role card explains a functional role: what has to be done, not a job title. One person can hold several of them, and one can be spread over several people.
The short answer
The semantic platform operator is the functional role that runs the knowledge graph environment: the digital twins and their pipelines, inference, the effective semantic view, the vector index and its incremental re-indexing. In Michael Iantosca’s white paper the key innovation is not a new database but a disciplined process inside the environment an organisation already runs; this role keeps that environment in step with the approved model version. It is usually held by documentation systems developers, not by application developers.
For: Documentation systems developers and technical project managers
Key points
- Runs the pipelines that fill the digital twins and configures inference.
- Defines and versions the effective semantic view the embeddings are made from.
- Re-indexes only what an approved change affects; candidates stay outside the official model.
Functional role
Semantic platform operator
The role in one sentence
The function that runs the knowledge graph environment — graphs, inference, the effective semantic view, the vector index and re-indexing — and keeps it in step with the approved model version.
Responsibilities
- Run the pipelines that fill the digital twins from the content, delivery and product sources.
- Configure inference and decide which inferences are materialized.
- Define and version the effective semantic view.
- Build the vector index and re-index incrementally after approved changes.
- Keep candidate changes in a test area outside the official model.
Decides
- The technical configuration of graphs, inference, view and index.
- Build schedules and which version runs where.
Does not decide
- Approving changes to the model.
- The content of the model or of the topics.
- Which agent answers are acceptable where safety is concerned.
Receives
- Approved model versions.
- Content, delivery and product data.
- Candidate changes to test.
Hands over
- Running graphs and index with version records.
- Test environments for candidates.
- Re-indexing records.
Typically held by
Reader roles of the pathway map that usually take on this function.
- Documentation systems developers
- The main holders of the function.
- Technical project managers
- Plan capacity, releases and dependencies.
On the map
- Digital twins per ontology· From knowledge graphs to adaptive semantics: the pathway map
Runs the graphs and their pipelines.
- Inference within the known vocabulary· From knowledge graphs to adaptive semantics: the pathway map
Configures the rules and materialization.
- Effective semantic view· From knowledge graphs to adaptive semantics: the pathway map
Defines and versions the view.
- Enriched vector index· From knowledge graphs to adaptive semantics: the pathway map
Builds the index.
- Incremental re-indexing· From knowledge graphs to adaptive semantics: the pathway map
Re-indexes after approved changes.
The context
What the role does
The operator keeps the machinery of the architecture running: one knowledge graph per ontology filled by its pipelines, the rules of inference and which inferences are materialized, the view presented to the vectorization and the index built from it. Every build records which model version and which view it used, so an answer can be traced back to the semantics behind it.
In the evolution cycle
The operator gives candidate changes a place outside the official model — for example separate named graphs — so they can be tested without touching it, hands the results over for evaluation and governance, and after approval re-embeds only the material the change affects. Nothing is promoted by the platform itself.
Questions readers ask next
- Does this role need a programmer?
- It needs someone who can configure and check the platform: pipelines, rules, views and builds. The work is about what runs and which version it uses, not about building an application.
- Why record the view version for every index build?
- Because the embeddings can only express what the view contained. Without the version, nobody can say which semantics an answer was based on.
Sources
- From knowledge graphs to adaptive semantics: evolving DITA and iiRDS digital twins for AI (white paper) — Michael Iantosca, 25 September 2026
- EvoOntology vs. iiRDS: understanding two divergent ontology paradigms — AI TechDoc Press (Saina Veigel), 23 September 2026
Review log and changes
Every context card is checked against its sources before it is published, and again whenever it changes; the date under the byline is the last review. Corrections (something was wrong) and additions (something was missing) are logged below with date and time (Berlin time). Typos, formatting and link fixes are not listed.
Reviewed
No corrections or additions since publication.